The study is specific to the following intervention approaches, as implemented in developing countries: (1) HlV voluntary counseling and testing, (2) peer education, (3) empowerment interventions, (4) condom social marketing, (5) needle exchange, (6) mass media, (7) partner notification, (8) abstinence-based interventions, (9) care as prevention, (10) family planning counseling for H1V-infected women, (11) structural and environmental interventions, (12) school-based sex education, (13) behavioral counseling, (14) diffusion & opinion leader interventions, (15) free condom distribution, and (16) dyadic couple-based interventions. We shall: (1) systematically review and archive the quantitative and qualitative evidence of the effectiveness of each intervention approach; (2) examine the content of the intervention programs studied to identify common "clusters" of overlapping intervention approaches. For quantitative studies identified with rigorous study designs the study will: (3) characterize salient social, environmental, and population-specific characteristics with secondary data sources in the settings where the studies were conducted, at the time that they were conducted, and; (4) meta-analyze the effectiveness of the interventions on HlV-related behavioral risk reduction both within and across the discrete intervention approaches, and via sub-group meta-analysis by intervention cluster, social, environmental, and population-specific characteristics. In sum, through this analysis we seek to gain a comprehensive understanding of the impact of HIV interventions in developing countries on behavioral risk reduction, and how the combination of intervention approaches, population targeting and social and environmental factors affect program success. Moreover, the analysis will help to identify strengths and weaknesses in the overall research agenda, which will help the field to avoid unnecessary duplication, and focus future research efforts in the areas of greatest need.